Automatic Characterization of the Parkinson Disease by Classifying the Kinematic Gait Patterns

Abstract: Traditionally, the Parkinson Disease (PD) is diagnosed and followed up by conventional clinical tests that are fully dependent on the expert experience. The diffuse boundary between normal and early parkinson stages and the high variability of gait patterns difficult any objective characte...

Full description

Autores:: Sarmiento Castillo, Fernanda Carolina

Tipo de recurso:

Fecha de publicación:: 2015

Institución:: Universidad Nacional de Colombia

Repositorio:: Universidad Nacional de Colombia

Idioma:: spa

Description
Summary:	Abstract: Traditionally, the Parkinson Disease (PD) is diagnosed and followed up by conventional clinical tests that are fully dependent on the expert experience. The diffuse boundary between normal and early parkinson stages and the high variability of gait patterns difficult any objective characterization of this disease. An automatic characterization of the PD is herein proposed by mixing up different measures of the ipsilateral coordination and spatiotemporal gait patterns which are then classified with a classical support vector machine (SVM). The strategy was evaluated in a population with parkinson and healthy control subjects, obtaining an average accuracy of 87% for the task of classification. The second approximation was developed under the rule that the ipsilateral coordination disturbances reflect the general motor control deficit described in PD, so that can be used in the objective characterization of the disease. Two ipsilateral coordination measures have been widely used in the identification of their patterns, the Relative Power Index (RPI) and the Point Estimates of Relative Phase (PERP). In this paper we look into the potential use of ipsilateral coordination patterns for the automatic characterization of the PD, therefore is proposed a comparative accuracy analysis of the RPI and PERP for the classification of the interest groups by a classical SVM. The strategy was evaluated in a population with parkinson (16 subjects) and healthy control subjects (7), obtaining an average accuracy of 94,6% and 82,1%, for PERP and RPI respectively.

Automatic Characterization of the Parkinson Disease by Classifying the Kinematic Gait Patterns

Publicaciones similares